Simons Institute Courses

Foundations of RL

Explore key reinforcement learning algorithms: TD learning, Q-Learning, and Natural Policy Gradient. Learn about finite-time performance bounds and their underlying principles.

• YouTube
• 1 hour 5 minutes
• Self-Paced
• Free Video

Adversarial Bandits: Theory and Algorithms

Explore adversarial multi-armed bandit theory, algorithms, and recent advances in data-dependent regret guarantees, structural bandits, and more. Compare full-information and bandit feedback in online learning.

• YouTube
• 1 hour 21 minutes
• Self-Paced
• Free Video

Stochastic Bandits: Foundations and Current Perspectives

Explore key algorithms for sequential learning in stochastic bandits, including Upper Confidence Bound and Thompson Sampling, and their adaptations for various constraints.

• YouTube
• 1 hour 9 minutes
• Self-Paced
• Free Video

Deep Learning in Structural Biology and Protein Design - How, Where, and Why

Explore deep learning applications in structural biology and protein design, focusing on AlphaFold2 and emerging methods for visualizing proteins and creating novel ones for therapeutic use.

• YouTube
• 1 hour 13 minutes
• Self-Paced
• Free Video

When Is Scale Enough?

Exploring performance predictability in deep learning as models and datasets scale up, focusing on large language models and investigating solvable problems through scaling alone.

• YouTube
• 1 hour 12 minutes
• Self-Paced
• Free Video

Distribution Shift as Underspecification, and What We Might Do About It

Explore strategies for addressing distribution shift in deep learning, focusing on underspecification and potential solutions for more robust AI systems.

• YouTube
• 1 hour 1 minute
• Self-Paced
• Free Video

Understanding the Robustness of Deep Learning

Exploring deep learning robustness, transfer learning dynamics, and overparameterization to gain insights into network functioning and improve model reliability under distribution shifts.

• YouTube
• 1 hour
• Self-Paced
• Free Video

Statistical Learning Theory and Neural Networks II

Explores advanced concepts in statistical learning theory and neural networks, focusing on optimization perspectives, gradient descent analysis, and the neural tangent kernel regime.

• YouTube
• 1 hour 2 minutes
• Self-Paced
• Free Video

Statistical Learning Theory and Neural Networks

Explore statistical learning theory for deep neural networks, covering uniform laws, Rademacher complexity, and optimization perspectives including gradient descent and neural tangent kernel regime.

• YouTube
• 59 minutes
• Self-Paced
• Free Video

Adversarial Examples in Deep Learning

Exploring adversarial examples in deep learning: design, defense strategies, and the relationship between model size and robustness against attacks. Insights from experts on implications and challenges.

• YouTube
• 1 hour 33 minutes
• Self-Paced
• Free Video

Causality in Machine Learning - Panel Discussion

Experts discuss interpretable machine learning in natural and social sciences, exploring causal inference and its applications across various fields.

• YouTube
• 49 minutes
• Self-Paced
• Free Video

Batch Arguments for NP and SNARGs for P from LWE

Explore advanced cryptographic concepts like batch arguments for NP and SNARGs for P, derived from Learning With Errors (LWE) assumptions, in this technical talk by Abhishek Jain.

• YouTube
• 1 hour 7 minutes
• Self-Paced
• Free Video

Optimal Broadcast Encryption and More from Evasive LWE

Explore advanced cryptographic techniques using evasive LWE, focusing on optimal broadcast encryption and its applications in secure communication systems.

• YouTube
• 46 minutes
• Self-Paced
• Free Video

New Quantum Cryptographic Primitives I

Explore cutting-edge quantum cryptography with Mark Zhandry, delving into novel primitives and their implications for secure communication in the quantum era.

• YouTube
• 1 hour 5 minutes
• Self-Paced
• Free Video

Quantum Advantage Without Structure

Explore cutting-edge research on quantum advantage, focusing on unstructured problems and their implications for quantum computing's potential superiority over classical methods.

• YouTube
• 1 hour 6 minutes
• Self-Paced
• Free Video